Cirrhosis afflicts nearly 3 million people in the US, and complications of cirrhosis are the 4th leading cause of death in individuals ages 45-65.[1] One complication is portopulmonary hypertension (PPHTN), defined as pulmonary arterial hypertension (PAH) in the setting of portal hypertension. PPHTN is found in ~6% of patients evaluated for liver transplantation (totaling almost 170,000 Americans with PPHTN) and significantly increases mortality. There is a very poor understanding of why PAH occurs in portal hypertension and essentially no mechanistic studies of PPHTN. In fact, there has never been a large NIH-funded multicenter study of PPHTN. To address this need, the two principal investigators have executed the only NIH-funded program of patient-oriented research in PPHTN (Pulmonary Vascular Complications of Liver Disease (PVCLD) Study Group)(R03 DK064103). We showed that variation in genes linked to estrogen signaling (aromatase and estrogen receptor alpha (ESR1)) and higher estradiol levels significantly increased the risk of PPHTN in patients with advanced liver disease. However, the mechanistic link remains unknown. Animal models and patients with PAH have increased progenitor cells in the bone marrow and circulating in blood (CEPs). Estradiol stimulates CEPs via the estrogen receptor alpha (ERalpha) and alpha blockade reduces CEP colony formation. Estradiol is metabolized by cytochrome P450 1B1 (CYP1B1) to metabolites which have both pro-angiogenic (16alpha-OHE{1}) and anti-angiogenic (2-OHE, 2-methoxyestradiol (2ME)) effects. A single nucleotide polymorphism (SNP) (rs1800440) in CYP1B1 and higher urinary 16alpha-OHE1 / 2-OHE ratio are both risk factors for heritable PAH. Therefore, we hypothesize that genetic determinants increase estrogen activity and influence estradiol metabolism, leading to increased levels of CEPs (indicating greater angiogenesis) and the development of PPHTN. We propose a cross-sectional analysis of a prospective cohort and a small ?proof-of-principle intervention study to answer three aims. We aim: 1) To determine whether variation in the rs1800440 SNP in CYP1B1 is associated with PPHTN, 2) To define the link between estrogen signaling, CEPs, and PPHTN, and 3) To determine if modulation of estrogen signaling using an ERalpha inhibitor (fulvestrant) influences CEPs and echocardiographic findings in PPHTN. The ultimate goal of this proposal is to identify mechanisms of and therapeutic targets for PPHTN. If CYP1B1 variants increase the risk of PPHTN, clinical trials using 2ME could be pursued. If we find that CEPs mediate the link between estrogen signaling and PPHTN, trials using anti-angiogenic compounds could be pursued. Finally, downregulation of CEPs by fulvestrant would identify ERalpha blockade as another approach. If these hypotheses are incorrect, we could use the genetic and biomarker data to examine alternative hypotheses.